| A | B |
|---|
| Where and when did this paper appear? | NEJM; 111402 |
| What was the steroid dosing regimen? | Dexamethasone 10mg IV Q6 hours; given either 15-20 minutes prior to antibiotics; or concurrently with antibiotics. |
| What was the antibiotic regimen in this trial? | Most patients got amoxicillin 2g IV Q 4 hours X 7 – 10 days; this regimen was based on known susceptibility patterns in the Netherlands; and was subject to change depending on Grams staining. |
| What was the general design of this study? | Prospective; randomized; placebo-controlled; double-blind trial |
| How many patients were enrolled? | 301; 157 to dexamethasone; 144 to placebo |
| What is the theoretical benefit of steroids in the setting of bacterial meningitis? | Animal studies suggest that bacterial lysis instigated by antibiotics leads to subarachnoid inflammation; neuronal death; bad brain; and other unfortunate stuff. |
| T or F: all patients in this study received steroids 15 – 20 min prior to antibiotic tx. | False. Study protocol was ammended after interim analysis to allow co-admin of steroids and abx—primarily to improve the rate of patient enrollment. |
| What was the primary outcome measure in this study? | Glasgow Outcome Score (not to be confused with the Glasgow Coma Score) |
| Describe the Glasgow Outcome Score | 1 – death; 2 – vegetative state; 3 – severe disability; 4 – moderate disability; 5 – mild or no disability. |
| What did the investigators consider to be a favorable primary outcome? | 4 or 5 on the Glasgow Outcome score. |
| What were the secondary outcome measures in this study? | Death; focal neurological abnormality (aphasia; CN palsy; monoparesis; hemiparesis; ataxia); hearing loss; GI bleed; fungal infection; herpes zoster; and hyperglycemia. |
| T or F: This study was manufacturer-supported | True: partly funded by a grant from NV Organon |
| How were study results reported? | As relative risk; which of course gives an inflated impression of the actual benefit |
| The study design of this paper incorporated a “last observation carried forward” methodology. Explain. | Patients who left the study after randomization were included in the final analysis; with their last observed characteristics “carried forward.” So a patient with a GOS of 4 who was lost to followup at 3 weeks was still used in the 8 week analysis; utilizing the GOS of 4. |
| T or F: Patients treated with dexamethasone had statistically higher rates of GI bleeding; herpes; fungal infection and hyperglycemia than patients in the placebo group. | False. More patients in the dex group had hyperglycemia (50 vs 37); herpes (6 vs 4) and fungus amongus (8 vs 5); but the differences were not statistically significant. |
| T or F: anbitiotic treatment was highly standardized in this study. | False. |
| Eight weeks after treatment; the ABSOLUTE risk reduction of an unfavorable outcome in dex-treated patients was 10%. What was the number-needed-to-treat? | 10. |
| T or F: Adjuvant treatment with dex had a significant beneficial impact on neurologic sequelae; including hearing loss | False. The authors argue because these sequelae are most often seen in severe disease; and because dex improves survival; dex may have paradoxically resulted in more sequelae. But they ALSO say that lower mortality did not result in increased rate of sequelae. Take your pick. |
| T or F: many of the patients in this study had subtle presentations of meningitis. | False. This study relied heavily on a population with relatively clear-cut cases of meningitis. Accordingly; there is some concern as to whether the results can be generalized to “all comers” with r/o meningitis; including those pts who aren’t as “sick” at presentation. |
| T or F: The authors note that there is data that a two-day steroid regimen is as effective as four-day; and recommend use of either regimen. | False. The authors recommend a four-day regimen. |
| T or F: on subgroup analysis; patients with pneumococcal meningitis demonstrated a significant benefit from dex; while those with Neisseria meningitis did not. | True. |
